How Does Grain Refinement Improve Steel Properties?

Larrin

Larrin

Knifemaker / Craftsman / Service Provider
Joined
Jan 17, 2004
Messages
4,844
I will try and reduce what takes a whole book and a coupe years metallurgy studies into a simple explanation:

Steel strength is a factor of how strong the bonds between the particles that make it up are. Knife Steel is composed of grains of martensite and balls/rods of carbides. The tighter the bonds, the stronger the steel. When a blade breaks is is because the bond between two grains reached the failure point and separated. This is what a crack is.

Let's use a wall as an example of our blade steel.
Coarse - If the wall is made of cinder blocks, that is equivalent to large grain. A crack runs along the grain boundaries, so it runs along the mortar joints between the bricks. It can easily progress along and turn at the next bric, propagating from the side it started at to the other side quickly. This take much less force to make the crack grow and the wall fail. Let's say that the wall is 10 blocks high. That means that the crack only has to move ten boundaries to make the wall fail.

If the wall is made of bricks, it is stronger but the grain boundaries are still easily disrupted. This is fairly coarse grain. If the wall is 40 bricks high, it takes more time and force, but it only has to make it those forty steps to fail.

Medium - If the wall is made up from concrete with 2"" to 3"" rocks in it, it is equivalent to medium grain. It is much stronger than the block wall, but a crack can still find boundaries to move along. Once a crack starts, it will jump from rock to rock and move to the other side. If the rocks are packed well, it may take over a hundred moves to cross the wall. Much better, but still not the best.

Fine - If the wall is cast from cement with sand and fine gravel in it it is very strong. A crack will be harder to start and not find an easy path across the wall. It may take a lot more force and severa thousand steps to make a crack go across the wall.

Now lets compare the durability of these walls to a knife blade. Coarse grain breaks easily. Hit the wall with a car backing up carelessly and it cracks easily. The blade fails and breaks in half for the same reason.
The medium grain wall of rock concrete will take a heavier blow fron[m a car put in drive instead of reverse to crack, but the crack will likely run from top to bottom. The knife blade will take more abuse, but once it reaches the failure point, the crack will move across the blade and destroy it.
The fine grain wall is really solid, it take a car moving at full speed to break it. Hit it with a small car at low speed and the car is what gets damaged. This is why we want fine grain.

The other reason is that the edge sharpness is determined by the grain size at the edge.
If the grains are large, the edge is less tough, and chipping occurs easier.
Medium is better, but you still have limitations.
Fine allows the edge to be taken down to super thin and still hold up without chipping.

Flex, bend, and break are also inter-related to "strength", but that is a different topic.
 
Last edited:
I would recommend my linked article for a distilling of a book down into something understandable.

The yield strength of the steel is not based on the strength of the bonds but based on the movement of dislocations in the material as I attempted to explain in the article.

Edge sharpness is not controlled by grain size because grains do not fall out in chunks. You can partially sharpen a grain. It only contributes to sharpness secondarily through an improvement to toughness or strength. So a stronger, tougher edge would be less likely to chip or wear so it would maintain a thin edge longer. But in terms of getting the sharpness grain size would not be an issue.
 
This is a difficult topic to introduce because dislocations are so important to material behavior but are basically absent from any discussion of steel metallurgy among knifemakers. So I am introducing a new and foreign topic which is difficult to understand without a lot of explanation, diagrams, and videos. However, it is difficult to talk about a long list of other topics without it. All strengthening mechanisms: martensitic strengthening, precipitation strengthening, grain refinement strenghening, cold rolling, solid solution strengthening, etc. are all related to dislocations. So there are a lot of topics I can't explain easily without introducing them. In materials courses there is a lot of math that is used to describe them which is obviously not very useful when writing for a lay audience so at least I don't have 10 equations in the article. Hopefully readers will ask questions so I know where gaps in understanding are.
 
I figured my super simplification was going to make Larrin sit up and say ... "Nooooo!"
I was lumping many other factors together to make a simple answer. As Larrin says, it is a much more complex subject worth reading more on. His blog site has great articles on metallurgy.

I was referring to how easily the edge will have chip form when I said, "Chunks falling off." I have found that the finer the grain, the finer the edge can be taken to and still survive in use. Part of this is in toughness and steel choice, but HT and grain size certainly seems to be a factor for me.
 
I was thinking the other night about how to possibly talk to customers about simple steels like AEB-L vs more complex ones like say S-30V, etc. What came to my mind was that yes, the S-30V will hold an edge longer, but you may end up taking off less metal when you EASILY resharpen AEB-L because you problem won't have to grind out as many little dings and chips. Kind of like when the Germans replaced steels like 115W8 with more standard "high speed" steel and such. Yeah, the high speed steel will stay sharp longer, but that assumes the you can keep the teeth from breaking off. ;)
 
Stacy E. Apelt - Bladesmith said:
I figured my super simplification was going to make Larrin sit up and say ... "Nooooo!"
I was lumping many other factors together to make a simple answer. As Larrin says, it is a much more complex subject worth reading more on. His blog site has great articles on metallurgy.

I was referring to how easily the edge will have chip form when I said, "Chunks falling off." I have found that the finer the grain, the finer the edge can be taken to and still survive in use. Part of this is in toughness and steel choice, but HT and grain size certainly seems to be a factor for me.
Click to expand...
Restating things is a good way to formulate things in your own mind, and doing so out loud allows those who know better to correct misunderstanding. That's part of why I am doing this website, as a way to research various topics that I have not had a chance to do so thoroughly, and to better internalize certain concepts. And if I'm wrong, it will be pointed out to me, I am sure.
 
I greatly appreciate you pointing out things that I state incorrectly ... or when I am just plain wrong. Having a chance to re-state them helps this forum have better info
 
I have made some modifications to the article in an attempt to improve clarity and understanding.
 
I thought it was pretty clear. Nice article Larrin. :thumbsup:

Something I wonder about is the correlation between strength/toughness and grain size. The simplified graphic implied it is a linear correlation, but I expect that is a simplification?

In my own experimentation I've seen a significant difference in the behavior of samples with coarse grain and fine grain, but I've struggled to find as much difference in fine vs ultra fine. At least with my crude methodology. Any thoughts on that relationship as the structure gets finer and finer?
 
Nathan the Machinist said:
I thought it was pretty clear. Nice article Larrin. :thumbsup:

Something I wonder about is the correlation between strength/toughness and grain size. The simplified graphic implied it is a linear correlation, but I expect that is a simplification?

In my own experimentation I've seen a significant difference in the behavior of samples with coarse grain and fine grain, but I've struggled to find as much difference in fine vs ultra fine. At least with my crude methodology. Any thoughts on that relationship as the structure gets finer and finer?
Click to expand...
It’s not linear, it’s the inverse square root of the grain size. There are websites and books that lay out the derivation of the Hall-Petch relationship that you can study if you wish. Also, in some materials it is not the grain boundaries that are the primary boundary for dislocation motion. For example, in martensitic steel it is the lath or packet boundaries of the martensite. However, the lath and packet size also correlates with grain size so the result is the same.

Hall-Petch breaks down at ultra fine grain sizes, around 10 nm, because there is a transition from dislocation pileups to grain boundary sliding; the grains are so tiny that pileups do not occur. The strength levels off or even decreases. However, 10nm grains cannot be made with practical methods so this is unlikely to affect any experiments you are referring to.

The toughness improvement was experimentally verified by measuring the ductile-to-brittle temperature which decreased with finer grains. Room temperature toughness relationships are not always as simple.
 
Two important things have often been left out of or discussions on the forum, grain boundaries and dislocations. Things happen in grain boundaries .They have higher energy levels and they tend to collect things , good or bad .Grains can come out in chunks if the grain boundary is brittle when carbides have collected there. Vanadium doesn't make grains smaller they slow down the growth of a grain by slowing grain boundary movement !
Electron beam microscopes have told us a lot .We can see how precipitates slow dislocation movement .We can see how formation of martensite creates many dislocations and those interlock to give strength ! We can even see the strain field around a precipitate. That in turn will disappear as we increase tempering temperature.
Older metallurgists rely on what we see while younger ones like Larrin rely with a more mathematical approach. Either way we both are trying to educate you -- 'cause metallurgy is fun !! :D
 
Stacy E. Apelt - Bladesmith said:
I will try and reduce what takes a whole book and a coupe years metallurgy studies into a simple explanation:
Click to expand...
Stacy E. Apelt - Bladesmith said:
I figured my super simplification was going to make Larrin sit up and say ... "Nooooo!"
Click to expand...
Stacy E. Apelt - Bladesmith said:
I greatly appreciate you pointing out things that I state incorrectly ... or when I am just plain wrong. Having a chance to re-state them helps this forum have better info
Click to expand...
Stacy, Why? Maybe I am out of line but I don't think this is appropriate... especially from a moderator. Larrin's thread didn't need any "simplification"... Especially, in the very first reply. If folks aren't getting it, Larrin, can field questions. If you figured your simplification would stir things up, why do it out of the gate? You are a moderator... I am moderator... for whatever reason some folks see us as more knowledgeable when they shouldn't. We just try to enforce the rules. We aren't chosen for our superior knowledge in the arena we frequent... we are chosen because we have shown the admins that we can keep the peace and are rational under difficult circumstances. Perhaps I'm being irrational, now for doing this on the public forum and in Larrin's thread(sorry Larrin) but I feel(aka, know) you are and have been overstepping your position I am quite tired of fielding complaints.... mostly, off-forum, as that's where everybody seems to be headed, these days.

And as a wise Canadian once said,

Sorry.

Rick
 
jdm61 said:
I was thinking the other night about how to possibly talk to customers about simple steels like AEB-L vs more complex ones like say S-30V, etc. What came to my mind was that yes, the S-30V will hold an edge longer, but you may end up taking off less metal when you EASILY resharpen AEB-L because you problem won't have to grind out as many little dings and chips. Kind of like when the Germans replaced steels like 115W8 with more standard "high speed" steel and such. Yeah, the high speed steel will stay sharp longer, but that assumes the you can keep the teeth from breaking off. ;)
Click to expand...

I have always liked the analogy of a steel edge being like a sand dune with boulders in it, some of the boulders stick out at the peak of the hill. The sand represents the steel and the boulders are the carbides.

If we think of the top peak of that dune wearing down the more boulders there are the less even the edge will be if we want the edge to form a perfect inverted V peak.
 
I’m smiling and clapping too and pretending to get it. Maybe someday.
Thanks for taking your time to educate though, it’s good to know this forum has accurate information.
 
You must log in or register to reply here.
Back
Top